On 26.02.2011 07:55, Tom Lane wrote:
> So we really need some refactoring here. I dislike adding another
> fundamental step to the ExecutorStart/ExecutorRun/ExecutorEnd sequence,
> but there may not be a better way.
Could you keep the sequence unchanged for non-EXPLAIN callers with some
refactoring? Add an exposed function like ExecutorFinishRun() that
Explain calls explicitly in the EXPLAIN ANALYZE case, and modify
ExecutorEnd to call it too, if it hasn't been called yet and the
explain-only flag isn't set.
-- Heikki Linnakangas EnterpriseDB http://www.enterprisedb.com